Dynamic risk management and control system and risk assessment early warning method for process industry

Abstract

The invention provides a dynamic risk management and control system and risk assessment early warning method for a process industry, and the system comprises a micro-service data design module, an engine configuration module, an AP I interaction module, a data configuration module, a task generation module, and a simulation module. Simulation, early warning and strategy generation of the whole production process of the process industry are realized by adopting a micro-service framework and a container dynamic arrangement technology, key scenes of the whole production process of the process industry are covered, specific scene application verification can be realized based on existing industrial control, safety, environmental protection, equipment management and the like, and data of various links and the like in the production process are acquired, so that the production efficiency is improved. The platform function and the use effect thereof are comprehensively evaluated in a specific scene, abnormal conditions are quickly identified and positioned, improvement measures are formulated and implemented, self-learning and self-adaptive capabilities are realized, early warning and management and control strategies can be flexibly adjusted, and the method has flexibility.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a dynamic risk management and control system and a risk assessment and early warning method for a process industry. Background Art

[0002] my country's process industries (such as chemical, petroleum, steel, etc.) are at a critical stage of intelligent upgrading and are faced with challenges such as improving the scientific nature of production and operation management and enhancing the accuracy of risk analysis to cope with complex process mechanisms, strong variable coupling and process dynamics. The existing risk management and control systems at home and abroad often show that the risk management and control systems are not stable in dealing with the complexity and dynamics of the process industry, and the risk perception, intelligent analysis and safety collaborative innovation levels of all elements of the production and operation of the process industry are not high. In particular, there is a lack of unified standards among various types of data, resulting in serious information islands and low efficiency in data integration and sharing.

[0003] Moreover, for the process industry, there is no research on the industrial software platform of the full-factor dynamic risk intelligent management and control system to achieve comprehensive dynamic risk perception assessment and intelligent management and control in the process industry, and to carry out application verification in the chemical, oil and gas, and metallurgical industries to achieve full-factor risk perception, intelligent early warning decision-making, and accurate identification of equipment and personnel behavior status, thereby comprehensively improving the dynamic risk intelligent management and control capabilities of the process industry.

[0004] In view of this, in view of the challenges of multiple security levels in process industry production and operation, complex dimensions and multi-source heterogeneity of all-factor data, how to design a multi-security collaborative protection technology architecture for production and operation based on distributed microservices can both support independent research of other sub-topics. Summary of the invention

[0005] In order to overcome the problems existing in the related technologies, the present disclosure provides a dynamic risk management and control system and a risk assessment and early warning method for the process industry to solve the technical problems in the related technologies.

[0006] One or more embodiments of this specification provide a dynamic risk management and control system for a process industry, including:

[0007] The microservice data design module is used to design the microservice database, classify microservices, and use containerization technology to provide a unified deployment and operating environment for microservices, as well as conversion adapters to assign unique identification information to each container; it is used to achieve unified management, scheduling, and security upgrades of components, and to call microservices according to user needs;

[0008] The engine configuration module is used to configure the engine according to the functional requirements of each microservice, and provide a combined engine scheduling service according to the test requirements;

[0009] API interaction module, used to implement cross-platform model calls and provide users with microservice interaction API interfaces;

[0010] Data configuration module, used to design unified configuration database and files, supporting dynamic adjustment;

[0011] A task generation module, used to generate a detection task based on the detection or analysis requirements analysis of the semantic recognition algorithm and send it to the message queue; and

[0012] The simulation module is used to obtain task information in the message queue, determine the required data, scenario data and detection process, determine the microservice and engine with the highest matching degree, call at least one corresponding microservice and engine, create a simulation scenario, filter and process the data required by each microservice, realize abnormal operating condition warning, fault diagnosis, traceability analysis, optimization decision, SIS verification, optimization and verification of data knowledge mining algorithms, and continuously optimize and adjust the abnormal warning strategy based on real-time production data and operation feedback. Finally, according to the detection results of each microservice, the analysis engine and strategy engine are used to optimize and adjust the strategy.

[0013] One or more embodiments of this specification provide a risk assessment and early warning method implemented by a dynamic risk management and control system in a process industry, including the following steps:

[0014] Step S1, obtaining historical data collected during the production process in the database through the system interface, including variables of the sensing part, logic part and execution part of the production site, or fusion data in the production process;

[0015] Step S2, determine the risk assessment microservice that can implement risk assessment and early warning, send historical data to the risk assessment microservice container address, and call the assessment service;

[0016] Step S3, the evaluation service determines the causal relationship between various factors in the production process through the Granger causal analysis method, describes the quantitative relationship between nodes by estimating the conditional probability density between various factors, and constructs a causal probability graph model of the system;

[0017] Step S4, inputting the risk data with risk labels and risk node labels, and normal data into the risk assessment model for training until the convergence condition is met, thereby obtaining a model for monitoring and early warning of production risks in the process industry, wherein the risk assessment model includes a feature extraction module and a risk assessment and classification module, the feature extraction module is a graph convolutional network, and the risk assessment and classification module is a spatiotemporal graph convolutional neural network or a graph convolutional network; and

[0018] Step S5, collect real-time data in the production process through the system data collection interface and send it to the risk assessment microservice container address to call the assessment service, so as to realize the monitoring and risk monitoring of each link in the production process, and realize early warning by predicting the risk fault data obtained by the system microservice.

[0019] The present disclosure provides a dynamic risk control system and risk assessment and early warning method for the process industry. The advantages are that the microservice framework and container dynamic orchestration technology are used to realize the simulation of the entire production process of the process industry, early warning and strategy generation, covering the key scenarios of the entire production process of the process industry. It can also realize specific scenario application verification based on existing industrial control, safety, environmental protection, equipment management, etc., and obtain data from various links in the production process. In a specific scenario, the platform function and its use effect are comprehensively evaluated, abnormal conditions are quickly identified and located, and improvement measures are formulated and implemented. The self-learning adaptive ability can flexibly adjust the early warning and control strategies, and it is flexible. Therefore, the dynamic risk control system has technical feasibility and wide applicability in application scenarios in the verification of the process industry; the platform comprehensively evaluates the platform function and its use effect in a specific scenario, quickly identifies and locates abnormal conditions, deeply analyzes the deficiencies of each model in the risk control process, and formulates and implements improvement measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 A block diagram of a dynamic risk management system for a process industry provided in one or more embodiments of this specification;

[0022] Figure 2 A schematic diagram of a dynamic risk management and control system architecture for a process industry provided for one or more embodiments of this specification; and

[0023] Figure 3 A schematic diagram of the structure of a computer device provided for one or more embodiments of this specification. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below in conjunction with the drawings in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0025] The present invention aims to achieve the risk management of typical process industry production processes, which involves multi-level risk identification and management, including functional safety, information security, physical safety, etc., by focusing on the core issues of all-factor multi-source heterogeneous data integration, complex risk assessment, and dynamic decision support in high-risk environments in process industry production operations. Figure 1 As shown, it is a schematic diagram of the dynamic risk management and control system for the process industry based on container microservices provided by this embodiment. The dynamic risk data in the process industry often has very complex structures and situations such as multi-source, heterogeneous, heterogeneous, multi-modal, nonlinear, and non-stationary. The full-factor data perception module built through the platform can perform pre-processing such as cleaning, correction, alignment, and fusion on the collected complex data to form standardized data that is easy to manage and use later, and then store it in the data middle platform database of the platform; the data middle platform relies on efficient distributed message queues to provide convenient data subscription and push functions; with the data middle platform as the support, this project designs a multi-security based on a distributed microservice system The collaborative protection architecture distributes personnel safety, process monitoring, equipment monitoring, environmental monitoring, and network monitoring to relatively independent microservices for subsequent modeling and analysis. At the same time, a risk collaboration microservice is established to exchange information, cross-validate, and coordinate management with the above five microservices. Each microservice subscribes to relevant data from the data center according to its own needs, implements data-driven analysis logic, and interacts with verification data sets, security rule sets, and algorithm tool sets to jointly realize dynamic risk intelligent management of the process industry, build a unified operating environment based on microservices and middleware, provide an operating environment and conversion adapters, realize unified management, scheduling, and security upgrades of components, and improve the stability and efficiency of the platform.

[0026] The present invention is described in detail below in conjunction with specific implementation methods and the accompanying drawings.

[0027] System Example

[0028] According to an embodiment of the present invention, a dynamic risk management and control system for a process industry is provided. Figure 1 As shown, it is a block diagram of a dynamic risk management and control system for a process industry provided in this embodiment. The dynamic risk management and control system for a process industry according to an embodiment of the present invention includes:

[0029] The microservice data design module is used to design the microservice database, classify microservices, and use containerization technology to provide a unified deployment and operating environment for microservices, as well as conversion adapters to assign unique identification information to each container; it is used to achieve unified management, scheduling and security upgrades of components, and call microservices according to user needs; among them, reference Figure 2 As shown, it is a schematic diagram of the dynamic risk management system architecture provided in this embodiment. Microservices are provided according to needs, such as risk assessment personnel safety detection, process detection, equipment detection, environment detection, network detection, SIS verification, data analysis, etc. In this embodiment, containerization technology can use Docker and Kubernetes technology.

[0030] In this embodiment, microservices and containerization are implemented through the following steps:

[0031] Each process industry is divided into multiple microservices according to the industry process. For example, the entire operation process of the process industry is divided into several links, such as personnel safety detection, process detection, equipment detection, environment detection, and network detection. The implementation process of each link is designed as a microservice, and then the multiple microservices are encapsulated through docker to generate images corresponding to each microservice. The images corresponding to each microservice are uploaded to the pre-created docker warehouse; in the docker warehouse, the images required for building and implementing the industrial process are obtained according to the preset orchestration tool; environment variables and startup commands are added to each microservice.

[0032] The engine configuration module is used to configure the engine according to the functional requirements of each microservice, including the algorithm engine, rule engine, scene rendering engine, process engine, scheduling engine, analysis engine, strategy engine, etc., among which, the combined engine scheduling service is provided according to the test requirements.

[0033] In one embodiment, the personnel safety microservice monitors operations and analyzes personnel behavior to prevent risks such as illegal operations and fatigue work; the process detection microservice analyzes and monitors parameters such as temperature and pressure in real time to detect process abnormalities; the equipment detection microservice predicts equipment failures based on equipment operating parameters, and provides early warnings and maintenance strategies; the environmental detection microservice monitors environmental hazards such as fireworks and dust, and analyzes and detects safety hazards; the network detection microservice continuously monitors network traffic, analyzes and prevents network intrusions and attacks.

[0034] The API interaction module is used to implement cross-platform model calls, provide users with a microservice interaction API interface, provide services to external applications through the API interface, or download the required microservices through the API interface and directly call them to the corresponding platform or system for reuse.

[0035] The data configuration module is used to design a unified configuration database and files, support dynamic adjustment, and achieve system availability through service discovery and compatibility verification mechanisms.

[0036] The task generation module is used to generate detection tasks based on the detection or analysis requirements analysis of the semantic recognition algorithm and send them to the message queue.

[0037] The simulation module is used to obtain task information in the message queue, determine the required data, scenario data and detection process, determine the microservice and engine with the highest matching degree, call at least one corresponding microservice and engine, and create a simulation scenario, filter and process the data required by each microservice, optimize and verify algorithms such as abnormal operating condition warning, fault diagnosis, source tracing analysis, optimization decision, SIS verification, data knowledge mining, etc., and continuously optimize and adjust the abnormal warning strategy based on real-time production data and operation feedback. Finally, optimize and adjust the strategy through the analysis engine and strategy engine according to the detection results of each microservice.

[0038] In this embodiment, the detection simulation module distributes the real-time data or historical data collected by the monitoring equipment to multiple containers providing microservices according to the required microservices. After each container filters out the data information to be processed that it needs, it processes and saves the processing results according to the preset industrial logic, and then determines the abnormal information through analysis engine, and determines the adjustment strategy through policy engine.

[0039] For example, the system provided in this embodiment can realize full-factor risk perception, intelligent early warning decision-making and personnel behavior recognition functions for the synthetic ammonia unit in the chemical industry. The platform can monitor the furnace status and personnel behavior in real time for the blast furnace process in the metallurgical industry, provide timely early warning and ensure safety. The platform combines equipment status, environmental changes and operating behavior for the catalytic cracking unit in the oil and gas industry to realize risk perception of the oil and gas chain.

[0040] The system provided in this embodiment adopts microservice framework and container dynamic orchestration technology to realize the simulation, early warning and strategy generation of the whole production process of process industry, covering the key scenarios of the whole production process of process industry, and can also realize specific scenario application verification based on existing industrial control, safety, environmental protection, equipment management, etc. By obtaining data of various links in the production process, the platform functions and their use effects are comprehensively evaluated in specific scenarios, abnormal conditions are quickly identified and located, improvement measures are formulated and implemented, and self-learning adaptive capabilities are realized. It can flexibly adjust early warning and control strategies and has flexibility. Therefore, the dynamic risk control system has technical feasibility and wide applicability in application scenarios in the verification of process industries; through the platform, the platform functions and their use effects are comprehensively evaluated in specific scenarios, abnormal conditions are quickly identified and located, and the deficiencies of each model in the risk control process are deeply analyzed, and improvement measures are formulated and implemented.

[0041] In this embodiment, the industrial process may include the occurrence stage, analysis stage, decision stage, solution stage, etc., and may also include abnormal operating condition warning, fault diagnosis, online SIS verification, fault tracing analysis, optimized decision-making, data knowledge mining, self-learning safety warning and other needs based on big data analysis.

[0042] Further, according to the above industrial process execution phase detection or analysis requirements, the task generation module is configured to perform the following steps:

[0043] Analysis submodule: Analyze the execution process of detection or analysis requirements based on semantic recognition algorithm, determine the logical relationship between each detection subtask and each subtask, and send each subtask to the message queue; for example, the user's requirement is to detect equipment detection anomalies, trace the source of the anomaly according to the anomaly, determine the anomaly factor, and provide a solution strategy. At the same time, the SIL level safety parameters of the process industry system must be verified online. Therefore, it includes four subtasks: equipment anomaly detection, anomaly tracing, strategy formulation, and SIL level online verification. Among them, equipment anomaly detection, anomaly tracing, and strategy formulation are logical sequences executed in sequence, and the SIL level online verification subtask can be executed independently of other subtasks. Therefore, the above subtasks can be split and logically confirmed through semantic analysis methods based on user test requirement information.

[0044] Furthermore, the detection simulation module specifically includes the following modules:

[0045] Task confirmation submodule: used to obtain subtasks in the message queue, determine the data, scene data and detection process required for the subtask, determine the microservice and engine with the highest matching degree, and call the corresponding microservice and engine.

[0046] Scenario construction submodule: Determine the test scenario according to the subtask, such as industrial environment, equipment operation, etc., call the preset scene rendering simulation engine to load the scene data into the scene rendering simulation engine to create the test scenario.

[0047] The detection submodule is used to implement the execution of each subtask based on the test scenario, microservices and detection process. Finally, according to the detection results of each microservice and subtask and the logical relationship between subtasks, the analysis engine and policy engine are used to optimize and adjust the strategy.

[0048] The setting of the detection simulation module in this embodiment can determine the execution priority of each task according to the logical relationship between each subtask. In the case of multiple tasks, the workflow can be dynamically created and updated in batches according to the task priority. This can also enable the process industry to integrate multiple security collaborative dynamic adaptation algorithms, self-learning safety early warning mechanisms and intelligent anomaly detection technologies, significantly improving the efficiency, intelligence and safety of risk intelligent management and control.

[0049] In this embodiment, a full-factor data perception module is also provided for pre-processing such as cleaning, correction, alignment and fusion of complex data collected such as system detection and verification results, to form standardized data that is easy to manage and use subsequently, and then stored in the system database.

[0050] In this embodiment, a microservice function management module is also provided for editing microservices, including constructing, editing, classifying, deleting and restoring basic online editing management functions of microservices; the model function management layer.

[0051] This system also studies the hybrid mode governance strategy for the risk coordination problem between microservices, and supports functions such as fault tolerance, current limiting, downgrade and fuse. Through middleware, service registration, discovery, load balancing, etc. are realized, and platform component services are uniformly managed to ensure efficient operation.

[0052] In this embodiment, a single component model is difficult to meet the needs in complex application scenarios, and multi-mode components face challenges in deployment, data conversion and collaborative work. A unified operating environment based on microservices and middleware is built, and an operating environment and conversion adapters are provided to achieve unified management, scheduling and security upgrades of components, thereby improving the stability and efficiency of the platform. Industry-verified data sets, security rule sets and multi-security collaborative algorithm tool sets are integrated to develop a microservice application environment that supports frameworks such as Spring Cloud and Service Mesh to ensure efficient and autonomous operation of the tool set.

[0053] In this embodiment, in order to solve the problem of flexible configuration of business processes and warning rules on the platform in complex production environments, a unified configuration database is designed to store system configurable parameters (such as thresholds, user permissions, and data rules) so that the system can adapt to different needs. Based on the dynamic configuration technology of business needs, warning rules, and business processes, a rule configuration model is proposed, covering dimensions such as events, actions, and priorities, and supporting customized warning rules and response measures. Through consistency protocols and distributed transaction processing, the consistency and reliability of the system in dynamic configuration are guaranteed, service hot plugging and seamless upgrades are achieved, and flexibility is improved.

[0054] This embodiment designs a unified configuration database and file, supports dynamic adjustment, and ensures high system availability through service discovery and compatibility verification mechanisms. A unified dynamic risk intelligent monitoring service bus is introduced to provide full life cycle management and control services such as risk assessment, early warning, and diagnosis, open up the data flow between security microservices and engines, avoid information islands, and achieve more efficient risk management.

[0055] In this embodiment, in the process industry, such as oil and gas, chemical industry, and steel, there is a large amount of data in the typical process of the process industry, including automatic data collected by sensors and equipment and manually collected data recorded by humans. However, there are small samples and imbalance problems in the data, with abundant safety data but missing accident data. When constructing a data set, it is necessary to balance the proportion of different types of data; it is necessary to solve the problem of small sample learning based on the inherent characteristics of process industry process data and effectively deal with the data imbalance phenomenon.

[0056] Therefore, this embodiment sets a data expansion microservice, and the specific working principle of the microservice is as follows:

[0057] Construct a conceptual kinetic model of the process industry process, and then, based on the constructed conceptual kinetic model of the process industry process and analyze industrial data, determine the data-driven method and the coupling relationship between various factors in the process, confirm and implement the adaptive dynamic update mechanism and update method of the risk model, and then based on the finite element numerical simulation method, use numerical simulation software to establish a kinetic numerical model of the process industry process and construct a mechanistic model of the kinetics of the process industry process, which can be used to simulate the cause-and-effect relationship between various factors in the production process; use the mechanistic model of the kinetics of the process industry process to expand the risk data to generate generated data that conforms to physical laws and has a similar distribution to the real signal, and use the generated data to expand the original training samples.

[0058] In this embodiment, based on the conceptual model of the dynamics of the process industry process, numerical simulation software is used to construct a 3D grid, input the process data of the historical process industry operation, and use the MODFLOW module of the numerical simulation software for calculation to obtain a preliminary dynamic unsteady flow numerical model of the process industry process; and adjust the numerical model parameters with the collected process data of the process industry to obtain the mechanism model of the dynamics of the process industry process.

[0059] In this embodiment, the data expansion microservices set up to solve the problems of small samples and data imbalance significantly alleviate the model bias and insufficient generalization ability caused by data limitations by generating high-quality synthetic data or migrating knowledge in related fields, thereby achieving comprehensive and in-depth mining and maximum value utilization of process industry production and operation data. This innovative framework not only solves the technical bottleneck in current data processing, but also provides a reference solution for data-driven decision-making in similar complex industrial scenarios.

[0060] In this embodiment, in order to further ensure the safety and stability of production operation, a SIL online verification microservice is also set up, in which the system also stores an information security library, an enterprise security rule library, etc., which specifically implements the following functions:

[0061] Determine the parameters required in the SIL verification process, the parameters including dynamic parameters and static parameters;

[0062] Based on the system of this embodiment, process industry system parameter data is collected in real time to realize process industry system simulation, determine dynamic parameter values ​​and static parameter values, and calculate the PFD and structural constraints of SIS in combination with the intelligent memory network algorithm to obtain the SIL verification level value of SIS, and then obtain a dynamic SIL verification curve based on the SIL verification level value.

[0063] In this embodiment, the required parameters involve variables of the sensing part, the logic part and the execution part. The time-varying variables in each part are defined as dynamic variables, and the remaining variables are defined as static variables. If the SIL verification results of the current SIS and the SIL verification results of the fitting prediction can meet the SIL verification requirements of the SIS, a safe operation management plan is provided, and the dynamic trend of the PFD and SFF is analyzed;

[0064] If the SIL verification results of the current SIS or the fitted predicted SIL verification results fail to meet the SIL verification requirements of the SIS, a system warning will be issued for the time nodes that fail to meet the SIL verification requirements, and a traceability analysis will be performed on the variables that fail to meet the requirements, providing a predictive maintenance strategy for the SIS.

[0065] The following is a specific case to illustrate the implementation process of risk assessment and early warning based on mechanism and data fusion implemented by the cost implementation example system:

[0066] First, the system obtains historical data collected during the production process in the database through an interface, including variables of the sensing part, logic part and execution part of the production site, or fused data in the production process; the system data stores multi-source data such as integrated personnel, process, environment, equipment, network, etc., such as operation data, operation information and sample data, etc. It can also include industry verification data sets used to verify the accuracy of model prediction results, risk evidence and grade library used by the model to determine risk levels, etc.;

[0067] The system determines the risk assessment microservice that can realize risk assessment and early warning, sends historical data to the risk assessment microservice container address, and calls the assessment service. The assessment service determines the causal relationship between various factors in the production process through the Granger causal analysis method, and describes the quantitative relationship between nodes by estimating the conditional probability density between various factors, and constructs the system's causal probability graph model; then the risk data with risk labels and risk node labels, as well as normal data are input into the risk assessment model for training until the convergence conditions are met, thereby obtaining a model for monitoring and early warning of production risks in process industries, wherein the risk assessment model includes a feature extraction module and a risk assessment and classification module, the feature extraction module is a graph convolutional network, and the risk assessment and classification module is a spatiotemporal graph convolutional neural network or a graph convolutional network.

[0068] The system then collects real-time data from the production process through the data collection interface and sends it to the risk assessment microservice container address to call the assessment service, thereby realizing monitoring and risk monitoring of each link in the production process, and the risk failure data obtained by the system microservice prediction is used for early warning.

[0069] In this embodiment, a fault tracing microservice is also provided for determining the root cause of the fault by reverse reasoning based on the risk fault data of other microservices. Based on the risk fault data after the execution of the microservice, the fault occurrence node is determined, and reverse reasoning is performed to the previous associated node according to the causal relationship represented by the arrow, and the conditional probability density is calculated to determine whether a fault occurs, until the root node in the probability graph model; if the probability density of the monitored sample is within the range of the upper and lower confidence limits, it is identified as a normal sample; otherwise, it is identified as a faulty sample.

[0070] Method Embodiment

[0071] According to an embodiment of the present invention, a risk assessment and early warning method based on the dynamic risk management and control system of the process industry is provided. Figure 3 As shown, it is a flow chart of the risk assessment and early warning method provided in this embodiment. The risk assessment and early warning method according to the embodiment of the present invention includes the following steps:

[0072] Step S1, obtaining historical data collected during the production process in the database through the system interface, including variables of the sensing part, logic part and execution part of the production site, or fused data in the production process; wherein the system data stores multi-source data such as integrated personnel, process, environment, equipment, network, etc., such as operation data, operation information and sample data, etc., and may also include industry verification data sets for verifying the accuracy of model prediction results, risk evidence and grade library used by the model to determine risk levels, etc.;

[0073] Step S2, determine the risk assessment microservice that can implement risk assessment and early warning, send historical data to the risk assessment microservice container address, and call the assessment service;

[0074] Step S3, the evaluation service determines the causal relationship between various factors in the production process through the Granger causal analysis method, describes the quantitative relationship between nodes by estimating the conditional probability density between various factors, and constructs a causal probability graph model of the system;

[0075] Step S4, input the risk data with risk labels and risk node labels, as well as normal data, into the risk assessment model for training until the convergence conditions are met, thereby obtaining a model for monitoring and warning of production risks in process industries, wherein the risk assessment model includes a feature extraction module and a risk assessment and classification module, the feature extraction module is a graph convolutional network, and the risk assessment and classification module is a spatiotemporal graph convolutional neural network or a graph convolutional network.

[0076] Step S5, collect real-time data in the production process through the system data collection interface and send it to the risk assessment microservice container address to call the assessment service, so as to realize the monitoring and risk monitoring of each link in the production process, and realize early warning by predicting the risk fault data obtained by the system microservice.

[0077] This embodiment also includes the following steps: based on the risk failure data after the execution of the microservice, the fault tracing microservice determines the node where the failure occurs, reversely infers to the previous associated node according to the causal relationship represented by the arrow, and calculates the conditional probability density to determine whether a failure occurs, until the root node in the probability graph model; if the probability density of the monitored sample is within the range of the upper and lower confidence limits, it is identified as a normal sample; otherwise, it is identified as a faulty sample.

[0078] In this embodiment, a fault tracing algorithm based on the fault tracing microservice using Bayesian network reasoning is used to trace the root cause of the fault by constructing a Bayesian network to guide maintenance.

[0079] This embodiment also includes a risk data expansion step, calling a data expansion microservice to construct a kinetic conceptual model of a process industry process, and then based on the constructed kinetic conceptual model of the process industry process and analyzing industrial data, determining the coupling relationship between the data-driven method and various factors in the process, confirming and implementing the risk model adaptive dynamic update mechanism and update method, and then based on the finite element numerical simulation method, using numerical simulation software to establish a kinetic numerical model of the process industry process and construct a kinetic mechanism model of the process industry process, which can be used to simulate the cause-effect relationship between various factors in the production process; using the kinetic mechanism model of the process industry process to expand the risk data, so as to generate generated data that conforms to physical laws and has a similar distribution to real signals, and using the generated data to expand the original training samples.

[0080] In this embodiment, based on the conceptual model of the dynamics of the process industry process, numerical simulation software is used to construct a 3D grid, input the process data of the historical process industry operation, and use the MODFLOW module of the numerical simulation software for calculation to obtain a preliminary dynamic unsteady flow numerical model of the process industry process; and adjust the numerical model parameters with the collected process data of the process industry to obtain the mechanism model of the dynamics of the process industry process.

[0081] In this embodiment, the data expansion microservices set up to solve the problems of small samples and data imbalance significantly alleviate the model bias and insufficient generalization ability caused by data limitations by generating high-quality synthetic data or migrating knowledge in related fields, thereby achieving comprehensive and in-depth mining and maximum value utilization of process industry production and operation data. This innovative framework not only solves the technical bottleneck in current data processing, but also provides a reference solution for data-driven decision-making in similar complex industrial scenarios.

[0082] The embodiment of the present invention is a method embodiment corresponding to the above-mentioned system embodiment. The specific operations of each process step can be understood by referring to the description of the system embodiment, which will not be repeated here.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and the contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

Claims

1. A dynamic risk management and control system for process industries, characterized by include: The microservice data design module is used to design the microservice database, classify microservices, and use containerization technology to provide a unified deployment and operating environment for microservices, as well as conversion adapters to assign unique identification information to each container; Used to achieve unified management, scheduling, and security upgrades of components, and to call microservices based on user needs; The engine configuration module is used to configure the engine according to the functional requirements of each microservice, and provide a combined engine scheduling service according to the test requirements; API interaction module, used to implement cross-platform model calls and provide users with microservice interaction API interfaces; Data configuration module, used to design unified configuration database and files, supporting dynamic adjustment; A task generation module, used to generate a detection task based on the detection or analysis requirements analysis of the semantic recognition algorithm and send it to the message queue; and The simulation module is used to obtain task information in the message queue, determine the required data, scenario data and detection process, determine the microservice and engine with the highest matching degree, call at least one corresponding microservice and engine, create a simulation scenario, filter and process the data required by each microservice, realize abnormal operating condition warning, fault diagnosis, traceability analysis, optimization decision, SIS verification, optimization and verification of data knowledge mining algorithms, and continuously optimize and adjust the abnormal warning strategy based on real-time production data and operation feedback. Finally, according to the detection results of each microservice, the analysis engine and strategy engine are used to optimize and adjust the strategy.

2. The dynamic risk management and control system for process industries as claimed in claim 1, characterized in that: The microservice data design module is used to split each process industry into multiple microservices according to the industry process, and then encapsulate the multiple microservices through containerization technology, generate images corresponding to each microservice, and upload the images corresponding to each microservice to a pre-created container warehouse; In the container warehouse, the image required for building and implementing the industrial process is obtained according to the preset orchestration tool; and environment variables and startup commands are added to each microservice.

3. The dynamic risk management and control system for process industries as claimed in claim 1, characterized in that: The detection simulation module distributes the real-time data or historical data collected by the monitoring equipment to multiple containers providing microservices according to the required microservices. After each container filters out the data information to be processed, it processes and saves the processing results according to the preset industrial logic. The analysis engine then analyzes and determines the abnormal information, and the policy engine determines the adjustment strategy.

4. The dynamic risk management and control system for process industries as claimed in claim 1, characterized in that: The detection simulation module specifically includes the following modules: Task confirmation submodule: used to obtain subtasks in the message queue, determine the data, scene data and detection process required for the subtask, determine the microservice and engine with the highest matching degree, and call the corresponding microservice and engine; Scenario construction submodule: determining the test scenario according to the subtask, calling a preset scene rendering simulation engine to create a test scenario by loading the scene data into the scene rendering simulation engine; The detection submodule is used to implement the execution of each subtask based on the test scenario, microservices and detection process. Finally, according to the detection results of each microservice and subtask and the logical relationship between subtasks, the analysis engine and policy engine are used to optimize and adjust the strategy.

5. The dynamic risk management and control system for process industries as claimed in claim 1, characterized in that: It also includes data expansion microservices, which realize the construction of a dynamic conceptual model of the process industry process. Then, based on the constructed dynamic conceptual model of the process industry process and analyzing the industrial data, the data-driven method and the coupling relationship between various factors in the process are determined, and the adaptive dynamic update mechanism and update method of the risk model are confirmed and implemented. Then, based on the finite element numerical simulation method, numerical simulation software is used to establish a dynamic numerical model of the process industry process and construct a mechanism model of the dynamics of the process industry process. The mechanism model of the dynamics of the process industry process is used to expand the risk data to generate generated data that conforms to physical laws and has a similar distribution to the real signal.

6. The dynamic risk management and control system for process industries as claimed in claim 1, characterized in that: It also includes a fault tracing microservice, which is used to reversely infer the root cause of the fault based on the risk fault data of other microservices; Based on the risk failure data after the execution of the microservice, the node where the failure occurs is determined, and the causal relationship represented by the arrow is reversely inferred to the previous associated node, and the conditional probability density is calculated to determine whether a failure has occurred until the root node in the probability graph model; if the probability density of the monitored sample is within the range of the upper and lower confidence limits, it is identified as a normal sample; otherwise, it is identified as a faulty sample.

7. The dynamic risk management and control system for process industries as claimed in claim 1, characterized in that: A SIL online verification microservice is also set up, and the receiving system collects the parameters required in the SIL verification process in real time, determines the dynamic parameter values ​​and the static parameter values, and calculates the PFD and structural constraints of the SIS in combination with the intelligent memory network algorithm to obtain the SIL verification level value of the SIS; and obtains a dynamic SIL verification curve based on the SIL verification level value; If the SIL verification results of the current SIS or the fitted predicted SIL verification results fail to meet the SIL verification requirements of the SIS, a system warning will be issued for the time nodes that fail to meet the SIL verification requirements, and a traceability analysis will be performed on the variables that fail to meet the requirements, and a corresponding predictive maintenance strategy for the SIS will be provided.

8. A risk assessment and early warning method implemented by a dynamic risk management and control system for a process industry according to any one of claims 1 to 7, characterized in that: Includes steps: Step S1, obtaining historical data collected during the production process in the database through the system interface, including variables of the sensing part, logic part and execution part of the production site, or fused data in the production process; Step S2, determine the risk assessment microservice that can implement risk assessment and early warning, send historical data to the risk assessment microservice container address, and call the assessment service; Step S3, the evaluation service determines the causal relationship between various factors in the production process through the Granger causal analysis method, describes the quantitative relationship between nodes by estimating the conditional probability density between various factors, and constructs a causal probability graph model of the system; Step S4, inputting the risk data with risk labels and risk node labels, as well as normal data, into the risk assessment model for training until the convergence condition is met, thereby obtaining a model for monitoring and early warning of production risks in the process industry, wherein the risk assessment model includes a feature extraction module and a risk assessment and classification module, the feature extraction module is a graph convolutional network, and the risk assessment and classification module is a spatiotemporal graph convolutional neural network or a graph convolutional network; as well as Step S5, collect real-time data in the production process through the system data collection interface and send it to the risk assessment microservice container address to call the assessment service, so as to realize the monitoring and risk monitoring of each link in the production process, and realize early warning by predicting the risk fault data obtained by the system microservice.

9. The risk assessment and early warning method according to claim 8, characterized in that: Also includes the steps: Based on the risk failure data after the execution of the microservice, the fault tracing microservice determines the node where the failure occurs, reversely infers to the previous associated node according to the causal relationship represented by the arrow, and calculates the conditional probability density to determine whether a failure has occurred until the root node in the probability graph model; if the probability density of the monitored sample is within the range of the upper and lower confidence limits, it is identified as a normal sample; otherwise, it is identified as a faulty sample.

10. The risk assessment and early warning method according to claim 8, characterized in that: Also includes the steps: Call the data expansion microservice to build a dynamic concept model of the process industry process. Then, based on the built dynamic concept model of the process industry process and analyze the industrial data, determine the data-driven method and the coupling relationship between various factors in the process, confirm and implement the adaptive dynamic update mechanism and update method of the risk model, and then use the numerical simulation software based on the finite element numerical simulation method to establish a dynamic numerical model of the process industry process and construct a dynamic mechanism model of the process industry process to simulate the cause-effect relationship between various factors in the production process. The mechanism model of the dynamics of process industry processes is used to expand risk data to generate generated data that conforms to physical laws and has a similar distribution to real signals, and the generated data is used to expand the original training samples.

Images